Search

To increase the knowledge about S. capitis in the neonatal setting, we conducted a nationwide 3-month survey in 38 neonatal intensive care units (NICUs) covering 56.6% of French NICU beds. We demonstrated 14.2% of S. capitis BSI (S.capBSI) among nosocomial BSIs. S.capBSI incidence rate was 0.59 per 1000 patient-days. A total of 55.0% of the S.capBSIs were late onset catheter-related BSIs. The S. capitis strains infected preterm babies (median gestational age 26 weeks, median birth weight 855 g). They were resistant to methicillin and aminoglycosides and belonged to the NRCS-A clone. Evolution was favorable in all but one case, following vancomycin treatment.

Objective: To study the association between preterm labor and bacterial vaginosis; in women with preterm labor, to determine whether vaginosis modifies the risk of preterm delivery., Study Design: Case-control study. We used Amsel's clinical criteria to test 102 patients hospitalized for preterm labor and 102 control patients for bacterial vaginosis., Results: Patients with preterm labor were diagnosed with bacterial vaginosis significantly more often (13.8%, 95% confidence interval (CI) (7.7-22.0) than control patients (0.0%, 95% CI (0.0-3.6)) (P<0.001). Among the former, the time elapsed to delivery was identical regardless of the patient's bacterial vaginosis status (elapsed time: 35.9 versus 37.1 days, rate of spontaneous preterm birth 42.9 versus 43.2%, not significant)., Conclusion: Bacterial vaginosis is associated with preterm labor. Nonetheless, it does not appear to predict preterm birth among these patients.

A prospective study was conducted in 3056 live-born infants delivered at the Jeanneade-Flandre maternity hospital of the Lille Teaching Hospital between January and August 1997. Clinical, laboratory test, and microbiological test findings were compared. A cohort of 1003 infants who remained in the maternity ward but were considered at increased risk of maternofetal infection (MFI) based on history and/or obstetrical criteria and/or neonatal criteria underwent routine collection of specimens including gastric fluid, auricular and anal swabs, amniotic fluid, and placental fragments. Microscopic examination of gastric fluid smears, the first result available to the clinician, was found to have 27.5% sensitivity (983 samples). Positive predictive value (PPV) was only 17.8% because of a high rate of colonization (16.8%), defined as absence of clinical symptoms and three peripheral specimens positive for the same organism. However, negative predictive value (NPV) was as high as 99.8% as a result of high sensitivity (97.8%) in the infected neonates. The gastric fluid smear was positive in 30% and 35% of neonates born to mothers with hyperpyrexia during early and late labor, respectively, and in 42% of neonates born to mothers with a history of group B streptococcus carriage during the pregnancy. Forty-two per cent of neonates with a history of fetal tachycardia had a positive gastric fluid smear. Diagnostic criteria for infection were three peripheral specimens positive for the same organism, C-reactive protein elevation, and/or one or more clinical signs suggestive of infection, and/or a positive central specimen (blood, CSF). The infection rate in infants who remained in the maternity ward was 1.6%. The most common causative organisms were group B streptococci. These findings illustrate the useful contribution of gastric fluid smears to the early diagnosis of MFI and confirm the predominant role of group B streptococci.

Patients severely neutropenic, when hospitalized, occasionally receive selective digestive decontamination, and the risk of vancomycin-resistant strain selection is a drawback since glycopeptide resistance is often associated with betalactam and aminoglycosid resistance. Bacterial translocation can lead to multiresistant bacterial sepsis. Eighteen Enterococcus faecium strains were collected from patients hospitalized in the leukemia unit of the Universitary Hospital of Lille (CHRU, Pr Bauters) between October 1992 and July 1997 and were studied. Nosocomial acquisition or endogenous origin were investigated to choose well-adapted prevention. All the vancomycin-resistant strains were shown by Polymerase Chain Reaction having the van A gene. The clonality of these strains was investigated by Pulsed-Field-Gel-Electrophoresis after Sma I restriction. Pulsotype analysis showed variable homology (52%-100%). Our results do not show evidence of patient-to-patient E. faecium transmission and suggest vancomycin-resistant strains were independently selected by antibiotic therapy from individual fecal flora. Except when epidemic events or happen, this strain isolation is more related to antibiotic prescription than misuse of isolation techniques.

An aminoglycoside in association with beta-lactam antibiotic are usually the most efficient treatment for Pseudomonas aeruginosa infections in cystic fibrosis patients. Tobramycin has the lower MICs than the other aminoglycosides to Pseudomonas aeruginosa. The aim of the work was to compare pharmacokinetics of tobramycin after once daily (15 mg/kg/day; 11 patients) or thrice daily dose (5 mg/kg/day; 9 patients) in combination ceftazidime (CAZ 200 mg/day in 3 inj. IVD) in sputum and sera for two weeks. No statistical difference in the serum concentration obtained in each group of patients was observed between the first and the 14th day. Serum concentrations were three fold higher when tobramycin was administered in once daily dose. Low through concentrations were quickly obtained, but they were slightly higher after thrice daily doses. Bronchial concentrations were 2 to 2.5 superior and near the critical concentration of tobramycin. The clinical efficacy were comparable in the two regimens.